The existing body of theory explaining how religion contributes to conflict is inadequate. This is because no existing theory is both sufficiently comprehensive, in that it describes most of the ways in which religion can become involved in conflict, and sufficiently dynamic, in that it takes into account the interrelationships between the various components of the theory. Also, few attempts are made to integrate such a theory into the general literature on conflict, and few such theories are tested, or even testable, in a large-n format (the use of statistical methodology to analyse a large number of cases in order to test theories and look for general trends). A more comprehensive and dynamic theory of religion and conflict is developed here based on the argument that religion has four basic functions in politics, society and conflict: to provide a value-laden belief system; to supply standards and criteria of behaviour based on that belief system; to organise adherents through its institutions; and to legitimate actors, actions and institutions. This theory is then integrated into Gurr's theory of ethnic conflict in a dynamic format that is testable in a large-n format.